This documentation is currently under development. Certain sections are not yet complete and will be added shortly.
For the complete documentation index, see llms.txt. This page is also available as Markdown.

Warmup

Preload all tenant configurations

get
/api/warmup

Sequentially loads configuration for all tenants to populate caches and avoid cold-start delays.

Purpose

  • Reduce First-Request Latency: Pre-populates caches before real traffic arrives

  • Prevent Request Pile-Up: Avoids concurrent configuration loads during app startup

  • Improve User Experience: Ensures fast response times from the first request

Behavior

  • Acquires an exclusive lock to prevent concurrent warmup operations

  • Returns HTTP 425 (Too Early) if another warmup is already in progress

  • Processes each tenant sequentially (not in parallel)

  • Continues warmup even if individual tenants fail (logs errors)

  • Respects cancellation tokens for graceful shutdown

Typical Usage

Should be called by:

  • Azure App Service warmup triggers

  • Kubernetes readiness probes

  • Deployment scripts

  • Health check systems

Performance Considerations

  • Duration depends on number of tenants

  • May take several seconds to minutes for large deployments

  • Does not block other API requests

Responses
200

Warmup completed successfully.

No content

get/api/warmup

No content

Last updated